Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Tesla, TSLA & the Investment World: the Perpetual Investors' Roundtable

This site may earn commission on affiliate links.
4680 represents a big risk to TSLA right now. They've been working on it for years and haven't got it figured out. At some point, you need to consider the possibility that it won't (and will never) work. If that happens, CT and Semi get pushed back a year or two ... not the end of the world but it'll hurt and we become reliant on Chinese battery makers for the foreseeable future.
They reduced their scrap bill by 40% which reduced their COGS by 25%. So, if their old scrap bill was 60% of COGS and they reduced it by 40% then their new scrap bill is 35% of COGS. (.4 x .6 = .24.) (.6 - .24 = .36)

I don't think the current 35% scrap bill is anywhere near an acceptable production number. Maybe for a chip maker, but shouldn't a battery maker be well South of 10%? Betting the CT and Semi production ramps on 4680 when they haven't solved manufacturing is risky.

Also, Drew presented the information in a way that sounded good but (IMO) hid the severity of the problem. Not cool.

I'd love to be wrong on this. I'm not an engineer or mathematician. I just want to understand the risks going forward.
 
Last edited:
4680 represents a big risk to TSLA right now. They've been working on it for years and haven't got it figured out. At some point, you need to consider the possibility that it won't (and will never) work. If that happens, CT and Semi get pushed back a year or two ... not the end of the world but it'll hurt and we become reliant on Chinese battery makers for the foreseeable future.
I recommend Cleanerwatt who has been keeping track of 4680 production progress, or lack there of.

On that note, I hate all these Tweeters/YouTubers who extrapolate to infinity. Some apparent robotics expert said there will be zillion bots by now 🤦🏻‍♂️. So someone wasted a question on asking Elon how many bots are there. I was ROFL at Elon’s bemused reply. 🤣

 
At least based on the specs displayed during the first cybertruck announcement, initial targeted length was 231.7 inches = 19.3 feet = 19 feet 3.7 inches. The new statement of "under 19 feet long" doesn't seem to be a huge change in dimensions, just an important one for those wanting to park in a 19-20 foot garage.

View attachment 958414
So Elon Musk is ~82" tall. Interesting!
 
Just because Dan's right about this statement doesn't mean he knows what he's talking about.
Dan is a lot better than Jonas bro.

Since it is the weekend:

Once, Dan liked this tweet & then quickly changed his mind 🤣


IMG_1994.jpeg
 
  • Funny
Reactions: wipster
I know I don’t post here much anymore. I don’t need to. I love Tesla so so much. I have become one with Tesla. I am Tesla.

I bought my mom a Tesla. My sister, brother, uncle, best friend, gf all own a Tesla because of my utter obsession with TESLA. I convinced an Uber driver to order one on my way to the shareholder meeting (he picked it up a few weeks later!). And I’m going to be picking up my 6th Tesla next week.

Yesterday I signed a contract for a 25kw Solar Roof with 3 Powerwalls. $172k before 30% tax credit. I bought 1250 shares in the low $100s (wtf?), and sold only some of them to buy this system. Incredible!!!

Thank you to the bears once a-frickin-gain. Between this and the cash secured puts I write you guys fund my Tesla lifestyle. Thank you BMW for selling me a complete lemon M2, I wouldn’t have purchased a Model S and TSLA in 2018 if it wasn’t for you. Thank you Amazon for helping an 18 year old me with 800 bucks make my first million, allowing me to invest in the greatest company ever

Thank you TMC and all of the original gangstas. You know who you are. I will never sell those OG shares.
 
I recommend Cleanerwatt who has been keeping track of 4680 production progress, or lack there of.

On that note, I hate all these Tweeters/YouTubers who extrapolate to infinity. Some apparent robotics expert said there will be zillion bots by now 🤦🏻‍♂️. So someone wasted a question on asking Elon how many bots are there. I was ROFL at Elon’s bemused reply. 🤣

Shame he is going downhill also, he had really good content, now he release videos with barely any content, making long pauses, repeating information many times, saying everything in the longest form possible, just for any new tiny piece of info to become a 10 min video

Haven't watched this last one yet, but I get sad when a really good channel goes on that path, his inside info and analysis from a few months ago on the 4680 progress were one of the best

The truth is, there isn't enough news or new info to have good Tesla content frequently, specially when you narrow it down to one area of interest
 
They reduced their scrap bill by 40% which reduced their COGS by 25%.
Relax Capt'n, this is a great outcome.

More importantly the cost of making cells for Cybertruck has hit their targets , we have no idea what that 40% yield improvement was...

It could be 20%->28% or 65%->91% both might give a COGS improvement of 25%, but it is slightly more likely when starting with a lower yield?

The alternative way to look at things is energy density and costs are already good enough, and they may have lots room for improvement in both areas.

Q2 production was an 80% improvement on Q1, with lots of new equipment going in and lines in the process of ramping, again room for improvement.

I watched the Cleanerwatt video he more or less repeated the info Drew provided, with an interpretation similar to a few of us here.. There was some number crunching on cell production volumes that was useful. Very few people who make YouTube videos regularly have additional inside information, mostly they start with what we already know. For those that don't already know, or want a well presented summary, they are useful.

In relation to providing additional information, Limiting Factor is by far the best Tesla related YouTube channel
 
Last edited:
Relax Capt'n, this is a great outcome.

More importantly the cost of making cells for Cybertruck has hit their targets , we have no idea what that 40% yield improvement was...

It could be 20%->28% or 65%->91% both might give a COGS improvement of 25%, but it is slightly more likely when starting with a lower yield?

The alternative way to look at things is energy density and costs are already good enough, and they may have lots room for improvement in both areas.

Q2 production was an 80% improvement on Q1, with lots of new equipment going in and lines in the process of ramping, again room for improvement.

I watched the Cleanerwatt video he more or less repeated the info Drew provided, with an interpretation similar to a few of us here.. There was some number crunching on cell production volumes that was useful. Very few people who make YouTube videos regularly have additional inside information, mostly they start with what we already know. For those that don't already know, or want a well presented summary, they are useful.

In relation to providing additional information, Limiting Factor is by far the best Tesla related YouTube channel
Slight correction:
It's a 40% reduction in scrap, not a 40% increase in yield.
But my brain isn't doing basic algebra without a notepad as it used to. So I'll skip on the calculation...
 
I wouldn't get too excited by that tweet. Green seems to live for the days when he can accuse tesla of screwing up. He also rushes out information based on features he sees in code that are not even active.
That's ad hominem.

There's no reason for Tesla not to comply with open-source licenses.

The company is so proud of their litigation team and is happy suing those who use their open-sourced patents (5 days ago) but still does not care others' open-source licenses (5 years ago)

In May 2018, it was reported that Tesla had for five or six years been using other people's copyrighted software unlawfully, specifically engaging in GPL violations. The Software Freedom Conservancy reportedly alerted Tesla to the issue repeatedly, but only in 2018 did Tesla begin to remedy its non-compliance with the software's license terms
How did Tesla behave in that front, since 2018?
 
Last edited:
If scrap bill is solely cells that fail at formation (versus scrap at all steps) then the yeild went up also which reduces amortized costs.

If we assume gross volume and gross cost per cell (pre formation) didn't change, then a 25% reduction in cell COGS on a 40% reduction in scrap implies that yeild improved from 55% to 73%, 18% absolute gain or 32% relative.

For simplicity: Cell gross cost $1, gross production 1000 cells, $1000 total gross cost
Yield 55%:
550 cells net, $450 scrap, $1000/550 = $1.82 per cell COGS
Yeild 73%:
730 cells net, $270 scrap, $1000/730 = $1.37 per cell
Delta scrap = 270/450 = 60% = 40% reduction
Delta COGS = 1.37/1.82 = 75% = 25% reduction

Reduction in scrap per cell: .37/.82 = 45% or a 55% reduction.
To get another 50% reduction in excess cost:
$1.18 per cell = $1000/847 = 85% yeild
 
If we assume gross volume and gross cost per cell (pre formation) didn't change, then a 25% reduction in cell COGS on a 40% reduction in scrap implies that yeild improved from 55% to 73%, 18% absolute gain or 32% relative.

I don't think so. Another way to look at this is the "40% reduction in scrap lead to a 25% reduction in COGS". Let's refer back to the Sep 2022 Battery Day Slide (pg 48) that explains the cost breakdown for cathode: (the most expensive component of the cell)


1690032601316.png


"PROCESSING" accounts for 35% of the cost of the cathode, whether it results in a useful cell or in scrap. If you can reduce scrap rate by 40%, you automatically reduce COGS as a percentage because output is now higher.

The COGS percentages would decrease something like this: -0.4*(0.35+0.25+0.05) = -26% or about what Drew said. Obviously, cathode isn't the only component of a cell, but it's the largest cost, and if processing is also about a third of other materials cost, then the above ratio also applies.

Cheers!
 
Last edited: